Load Environment Variables Conditionally | Generated by AI
Fixed. Here’s what I changed in ~/.hermes/plugins/note/init.py:
python
Load ww's .env so LLM calls work even without shell-level exports
try:
from dotenv import load_dotenv
load_dotenv(Path.home() / "projects" / "ww" / ".env", override=False)
except ImportError:
pass
Key detail: override=False — if you already have MODEL set in your shell, it won't be overwritten. It only fills in missing vars from ww's .env.
The try/except ImportError handles the edge case where python-dotenv isn't installed (it almost certainly is, but defensive).
Try /note again — it should work now.